The tissue sample may result from a needle biopsy (sometimes referred to as tru-cut biopsy), fine needle aspirate or an explant. A needle biopsy is a common medical test involving the removal, with the aid of a hollow needle, of a representative sample of cells or tissues from a living subject for examination to determine the presence or extent of a disease. Fine needle aspiration (FNA) is a diagnostic procedure sometimes used to investigate superficial (just under the skin) lumps or masses or hollow organs. According to this technique, a thin hollow needle is inserted into the mass for sampling of cells that, after being stained, will be examined under a microscope. A biological sample harvested from a piece or pieces of tissue is called an “explant.” The explant and/or biopsy sample is subjected to treatments, such as fixation, sectioning and staining, and then examined by a pathologist. The tissue sample can also be analyzed chemically. For the sake of brevity, the term “biopsy” will be used throughout this application with the understanding that it refers to tru-cut biopsies as well as to fine needle aspirates and free-form explants.
Patterned biopsies play a major role in the early detection of various types of cancer (i.e., prostate, breast, thyroid, skin, intestine, lung, stomach, etc.). Maintaining site-specific information regarding individual biopsy cores is of critical importance. While individual processing of biopsies by conventional methods is prohibitively expensive, current methods of parallel processing (tissue microarrays, color coding, multi-compartment cassettes, etc.) are either not accurate, cumbersome, or both.
A standard biopsy method uses a needle to remove a biopsy sample for examination. A tru-cut needle biopsy removes small but solid samples of tissue using a hollow “core” needle. In a core biopsy, a small cylindrical sample of tissue is removed preserving the histological architecture of the tissue's cells. This is important when dealing with biopsies (e.g., prostatic, breast) where one has to report the percentage of malignant tissue. The samples are small, and fragile, and tend to curl during processing, making preparation of the sample for examination difficult.
In the United States, there are approximately 1 million patients requiring biopsies for early diagnosis of prostate cancer. On average, 10 to 21 biopsy samples are taken per patient with suspicion of prostate cancer and their precise location has to be recorded. In some cases “prostate mapping”/saturation biopsies are performed (up to 60 to 80 biopsy samples per patient). Utilizing the standard biopsy technique (in which one, or maximum two, biopsies are processed at a time) results in a huge number of paraffin blocks, requiring a large number of sections and slides, high expenditures in terms of consumables, manpower for processing and time spent by the pathologists for interpreting the slides. Moreover, in order to provide the required 3-D reconstruction analysis of malignancy, it is very important to record and map the location of the sections relative to the original biopsy sample.
Traditionally, when dealing with biopsy samples, the recommended approach is to process the samples by embedding individually in a supporting material such as a paraffin block. The paraffin blocks containing the individual samples are then sectioned with a microtome to produce thin sections that will be placed on a microscope slide, stained as needed and examined under a microscope. When sectioning the paraffin blocks, one runs the risk of not intercepting the tissue sample and/or losing too much of the sample before a (quasi) complete section is produced. For example, when dealing with prostate biopsies, current recommendations require three different sections (“step” sections, or “levels”) taken at approximately 50 microns apart. Usually, two to three additional sections (unstained) are saved at every step/level for further staining, if required by the pathologist. Sufficient material should remain in the block for further study as well as for archiving, litigation, etc. Because of the scarcity of the material in the sample, only adhesive-coated slides (i.e., 10 to 20 times more expensive than regular variety) are employed, to minimize the risk of accidentally losing the sections during staining.
In recent years, trans-rectal ultrasound-guided (TRUS) systematic needle biopsy has emerged as a new gold standard in prostate cancer diagnosis, to such an extent that statistical performance values (sensitivity, specificity, positive and negative predictive values) of all other diagnostic tests, like digital rectal examination (DRE) or prostate-specific antigen (PSA) are computed according to the outcome of biopsy examination. A computerized model of the prostate, including mapped sections from 159 whole-mount radical prostatectomy specimens, subjected to systematic histopathology examination showed that the six-core biopsy technique fails to identify 26.8% of the tumors. Even on repeated examination, this method resulted in failure of tumor identification in 27 of 118 prostate cancer patients, equivalent to 23% of total cases demonstrated following radical prostatectomy. From a plethora of recent studies comparing results of different biopsy strategies regarding the number and location of cores, the emerging consensus, based on clinical trials, ex vivo biopsy approaches, as well as various computer simulations and mathematical models, is to take at least ten biopsy cores, focusing the biopsies laterally at the base, mid-gland, and apex of the prostate, with mid-lobar biopsy cores at the base and apex, and adjusting the number of cores taken according to prostate volume and age of the patient. Some investigators advocate even more aggressive biopsy schemes, with more than 12 cores, up to a saturation biopsy (>20 cores), especially on repeat biopsies, reporting even higher cancer detection rates. However, despite the obvious need for multiple biopsy cores per patient, due to high expenses and limited resources, the number of biopsies taken and sections prepared per patient is typically reduced to the minimum required for an acceptable precision of the diagnosis and/or the maximum resources available within an institution. Thus, a system for rapid, cost-effective preparation and analysis of multiple cores at multiple levels is needed.
An array is an organized fashion of multiple tissue and/or cell samples that can be used in various histological techniques including topographical staining, (immuno)histochemistry, immuno-fluorescence, and in situ hybridization. Tissue and cell arrays are powerful tools because they allow simultaneous screening of numerous tissue or cell samples. The value of this type of technology is that testing can be done on many samples in a timely manner with consistency. This allows for high-throughput histological screening or analysis.
There are a number of methods that describe the construction of tissue or cell arrays including: the “sausage” method (see Battifora, “The multitumor (sausage) tissue block: novel method for immunohistochemical antibody testing” (1986) Lab. Invest. 55:244; and U.S. Pat. No. 4,820,504); paraffin-to-paraffin transfer methods (see Kononen et al., “Tissue microarrays for high throughput molecular profiling of tumor specimens” (1998) Nature Medicine 4(7):844-847); the “honeycomb” method (see K. Petrosyan and M. F. Press, “Multispecimen tissue blocks in pathology: an improved technique of preparation” (1997) Lab. Invest. 77(5):541-542); and the use of liver as a recipient matrix for the array (see Musat-Marcu et al., “Inhibition of apoptosis after ischemia-reperfusion in rat myocardium by cycloheximide” (1999) J. Mol. Cell. Cardiol. 31:1073-1082). Also see U.S. Pat. No. 4,647,543; Miller and Groothius, A.J.C.P. 96:228-232; Sundbland, A.J.C.P 102:192-193; Patent Application No. WO1999IUS9912537; Patent Application No. PCT/US99/04000; Patent Application No. WO1999/WO0004001 and U.S. patent application Ser. No. 1987000110818.
The “sausage” technique involves combining multiple tissue samples from a deparaffinized block into a single composite “sausage” held together with a wrapper of intestinal casing. The multiple tissue sausages are re-paraffinized, sectioned and then mounted on slides. This procedure allows hundreds of tissue samples to be tested simultaneously. Even though this approach is valuable, it has a number of inherent disadvantages. For instance, the technical effort and time required to prepare the composite sausage causes difficulties. In addition, the need to de-paraffinize and re-paraffinize the tissue samples could lead to a loss of antigens. There are problems in maintaining the spatial relationships among the different tissue samples and working with small specimens such as cells, because of the flexible nature of the intestinal casing.
There have been a number of other paraffin-to-paraffin or double-embedding techniques that have evolved to fix some of the problems with the “sausage” technique (see U.S. Pat. Nos. 4,914,022 and 5,002,377). A common multi-specimen technique involves the preparation of standard paraffin blocks of tissue specimens where core samples are then removed from these blocks and re-embedded into a recipient paraffin block to create the tissue array (see U.S. Pat. No. 4,914,022). This technique can be used to test multiple tissue samples from multiple sources at the same time. In addition, it is possible to take a tissue chip, which is a thin section of the tissue array, and parallel process a number of samples at the same time with a variety of stains or molecular markers. The problem with this technique is that in order to prepare the tissue chips, custom-built equipment is required, which includes a computer-controlled micro-stage. In addition, the recipient paraffin block cannot be cut unless the adhesive tape technique is employed. The adhesive tape technique is expensive and time-consuming, and because the cut sections require special treatments before staining, there is a risk of compromising the accuracy of many applications. Another problem with paraffin-to-paraffin techniques is that the recipient paraffin block cannot cut efficiently and, thus, a number of sections are lost. Also, serial sectioning of the recipient block to produce ribbons is virtually impossible. Another limitation of this technique is that the tissue samples must be paraffinized and then re-paraffinized, thus, it is not possible to use fresh samples to create the microarray and antigens might be lost in the process. This technique can also be used to construct cell arrays; however, the same disadvantages exist as found when constructing tissue arrays (see M. Cottler-Fox and C. H. Fox, J. Infect. Dis. (1991), 164:1239-1240).
The “honeycomb” technique to create tissue arrays has the advantage of being able to employ fresh or fixed tissue without prior embedding. This technique uses a multi-chambered mold (“honeycomb”) made with plastowax. Small tissue specimens are placed in the equal-sized spaces of the mold, and then the molds are embedded with PARAPLAST®. The multi-specimen tissue blocks are processed, sectioned and stained using conventional methods. The problem with this technique is that precise orientation of the individual samples is not possible. As a result, its applications are very limited and, as such, it is not widely used. Like all previous methods, it requires infiltration followed by embedding, for example, in paraffin.
Liver tissue has previously been proposed as a support for facilitating sectioning of tissue samples (see, e.g., Manfred Gabe, “Histological Techniques,” p. 125 (Springer-Verlag 1976), and Musat-Marcu et al., “Inhibition of apoptosis after ischemia-reperfusion in rat myocardium by cycloheximide,” J. Mol. Cell. Cardiol. 31:1073-1082 (1999). Tissue and cell arrays can be constructed using liver as the recipient matrix. This technique works for fixed, fresh, or paraffinized tissue or cell samples. The problems with this technique include the low efficiency in generating the required matrices and the occasional hidden “defects” in the liver matrix, such as biliary ducts, blood vessels and collagen septa that results in a loss of samples. In addition, there is possible cross-reactivity between the matrix and biopsy samples.
Thus, although tissue arrays and cell arrays are powerful tools to allow simultaneous screening of numerous tissue or cell samples, there are no simple methods for creating arrays using core biopsy sample or free-form explants.